首页>
外国专利>
MACHINE LEARNING METHODS AND APPARATUS RELATED TO PREDICTING MOTIONS OF OBJECTS IN A ROBOT'S ENVIRONMENT BASED ON IMAGES CAPTURING THE OBJECTS AND BASED ON PARAMETERS FOR FUTURE ROBOT MOVEMENT IN THE ENVIRONMENT
MACHINE LEARNING METHODS AND APPARATUS RELATED TO PREDICTING MOTIONS OF OBJECTS IN A ROBOT'S ENVIRONMENT BASED ON IMAGES CAPTURING THE OBJECTS AND BASED ON PARAMETERS FOR FUTURE ROBOT MOVEMENT IN THE ENVIRONMENT
Some implementations of the present specification generally refer to a deep machine learning method and apparatus for predicting (if any) motion(s) that will occur for an object(s) in the robot's environment in response to a specific movement of the robot within the environment. It is about. Some implementations use a deep neural network model to predict at least one transformation (if any) of an image of the robot's environment, which will occur as a result of implementing at least some of the robot's specific movement within the environment. It's about training. The trained deep neural network model can predict transformation based on an input including an image and a group of robot motion parameters that define a part of a specific motion.
展开▼